Skip to content

Latest commit

 

History

History
 
 

seaformer

SeaFormer: Squeeze-Enhanced Axial Transformer For Mobile Semantic Segmentation

Reference

Qiang Wan, Zilong Huang, Jiachen Lu, Gang Yu and Li Zhang. "SeaFormer: Squeeze-Enhanced Axial Transformer For Mobile Semantic Segmentation." Published as a conference paper at ICLR 2023, arXiv:2301.13156v4.

Performance

ADE20k

Model size Head Resolution Training Iters mIoU (slice) Links
SeaFormer tiny LightHead 512x512 160000 34.58 model | log | vdl
SeaFormer small LightHead 512x512 160000 38.73 model | log | vdl
SeaFormer base LightHead 512x512 160000 40.92 model | log | vdl
SeaFormer large LightHead 512x512 160000 43.66 model | log | vdl